Granted Patent B2

US 12,301,885 · App. 18/116,804 · Granted May 13, 2025

Inventor: Sachin G. Deshpande (Camas, WA)

Assignee: Sharp Kabushiki Kaisha

H04N19/80G06T9/002H04N19/70

Abstract

Claims (17)

1. A method of performing neural network filtering for video data, the method comprising:

receiving a neural network post-filter characteristics message specifying a neural network post processing filter;

parsing a first syntax element in the neural network post-filter characteristics message, wherein the first syntax element plus one specifies a denominator of a resampling ratio of a width of a picture generated by the neural network post processing filter relative to a cropped width;

parsing a second syntax element in the neural network post-filter characteristics message, wherein the second syntax element plus one specifies a numerator of the resampling ratio of the width of the picture generated by the neural network post processing filter relative to the cropped width;

parsing a third syntax element in the neural network post-filter characteristics message, wherein the third syntax element plus one specifies a denominator of a resampling ratio of a height of a picture generated by the neural network post processing filter relative to a cropped height;

parsing a fourth syntax element in the neural network post-filter characteristics message, wherein the fourth syntax element plus one specifies a numerator of the resampling ratio of the height of the picture generated by the neural network post processing filter relative to the cropped height;

calculating a width of a luma sample array of the picture resulting from applying the neural network post processing filter by multiplying the cropped width by the resampling ratio of the width; and

calculating a height of the luma sample array of the picture resulting from applying the neural network post processing filter by multiplying the cropped height by the resampling ratio of the height.

2. A device comprising one or more processors configured to:

receive a neural network post-filter characteristics message specifying a neural network post processing filter;

parse a first syntax element from in the neural network post-filter characteristics message, wherein the first syntax element plus one specifies a denominator of a resampling ratio of a width of a picture generated by the neural network post processing filter relative to a cropped width;

parse a second syntax element in the neural network post-filter characteristics message, wherein the second syntax element plus one specifies a numerator of the resampling ratio of the width of the picture generated by the neural network post processing filter relative to the cropped width;

parse a third syntax element in the neural network post-filter characteristics message, wherein the third syntax element plus one specifies a denominator of a resampling ratio of a height of a picture generated by the neural network post processing filter relative to a cropped height;

parse a fourth syntax element in the neural network post-filter characteristics message, wherein the fourth syntax element plus one specifies a numerator of the resampling ratio of the height of the picture generated by the neural network post processing filter relative to the cropped height;

calculate a width of a luma sample array of the picture resulting from applying the neural network post processing filter by multiplying the cropped width by the resampling ratio of the width; and

calculate a height of the luma sample array of the picture resulting from applying the neural network post processing filter by multiplying the cropped height by the resampling ratio of the height.

3. The device of claim 2 , wherein the device includes a video decoder.

Assignments (1)

Continuity (1)

References Cited (18)